In this question that was asked today the OP wrote that $$\begin{align}\sum\limits_{n=1}^{∞}[\arctan(2n+1)-\arctan(2n-1)]&=\arctan\infty-\arctan 1\\ &=\frac{\pi}{2}-\frac{\pi}{4}\\&=\frac{\pi}{4} \end{align}$$
I don't really understand why there is a $\arctan\infty$ term. Surely every single term other than $-\arctan 1$ has been cancelled out? Now I know that I am wrong, as obviously the value of the summation cannot be negative, but I'm not sure where I am wrong.
I have thought of considering the similar finite series, $$\sum\limits_{n=1}^k[\arctan(2n+1)-\arctan(2n-1)]=\arctan(2k+1)-\arctan1$$ and letting $k\to\infty$, so that $$\lim_{k\to\infty}\sum\limits_{n=1}^k[\arctan(2n+1)-\arctan(2n-1)]=\arctan\infty-\arctan1$$ as required, but I still can't see why all the terms other than $-\arctan1$ don't cancel out, as the upper limit actually is $\infty$.
Thank you for your help.
A few intermediate steps might be helpful.
Comment:
In (1) we apply telescoping.
In (2) we use $\lim_{N\to\infty}\arctan(N)=\frac{\pi}{2}$ and $\lim_{N\to\infty} a=a$.